@softspark/ai-toolkit 1.0.0 → 1.1.0

package/CHANGELOG.md

@@ -7,6 +7,19 @@ Versioning follows [Semantic Versioning](https://semver.org/).
 
 ---
 
+## [1.1.0] - 2026-04-02
+
+### Added
+- **Agent Teams auto-enabled** — `CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS=1` is now automatically set in `~/.claude/settings.json` via `env` field during `install` / `update`. No manual `export` needed.
+- 2 new install tests: env var injection + user env var preservation
+
+### Fixed
+- README test badge count: 308 → 310
+
+[1.1.0]: https://github.com/softspark/ai-toolkit/releases/tag/v1.1.0
+
+---
+
 ## [1.0.0] - 2026-04-02
 
 ### Added
@@ -46,7 +59,7 @@ Versioning follows [Semantic Versioning](https://semver.org/).
 - CI: auto-regenerate `AGENTS.md` and `llms.txt` on push to main; publish on tags
 - **DRY refactoring**: `scripts/_common.py` shared library for all generators and CLI scripts; `app/hooks/_profile-check.sh` for 9 hooks; CLI uses data-driven `GENERATORS` map
 - All hooks standardized to `#!/usr/bin/env bash` shebang
-- 308 tests across 16 test files
+- 310 tests across 16 test files
 - **Iron Law enforcement** in TDD, debugging, and verification skills — anti-rationalization tables prevent agents from skipping test-first, root-cause analysis, or claiming completion without evidence
 - **`/subagent-development`** — 2-stage review workflow: dispatch implementer → spec compliance review → code quality review per task, with 4-status protocol (DONE, DONE_WITH_CONCERNS, NEEDS_CONTEXT, BLOCKED) and prompt templates
 - **`/repeat`** — Ralph Wiggum autonomous loop pattern with safety controls: max iterations (default 5), circuit breaker (3 consecutive failures → halt), min interval (1 minute), exit detection, stats logging

package/README.md

@@ -6,7 +6,7 @@
 [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](LICENSE)
 [![Skills](https://img.shields.io/badge/skills-85-brightgreen)](app/skills/)
 [![Agents](https://img.shields.io/badge/agents-47-blue)](app/agents/)
-[![Tests](https://img.shields.io/badge/tests-308%20passing-success)](tests/)
+[![Tests](https://img.shields.io/badge/tests-310%20passing-success)](tests/)
 
 ---
 
@@ -416,11 +416,7 @@ All packs have `status: experimental`. Each has a `plugin.json` manifest and `RE
 
 ## Agent Teams
 
-Native support for `CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS=1`:
-
-```bash
-export CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS=1
-```
+Native support for `CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS=1` — automatically enabled during `ai-toolkit install` / `update` via `env` in `~/.claude/settings.json`.
 
 Pre-configured team presets via `/teams`:
 

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@softspark/ai-toolkit",
-  "version": "1.0.0",
+  "version": "1.1.0",
   "description": "Professional-grade AI coding toolkit: 85 skills, 47 agents, multi-platform support (Claude, Cursor, Windsurf, Copilot, Gemini, Cline, Roo Code, Aider), machine-enforced safety constitution, Iron Law enforcement, persistent memory, visual brainstorming, expanded lifecycle hooks, 11 plugin packs, and benchmark tooling.",
   "keywords": [
     "claude",

package/scripts/install_steps/hooks.py

@@ -25,6 +25,7 @@ def install_hooks(claude_dir: Path, hooks_scripts_dir: Path,
         output_styles_dir = app_dir / "output-styles"
         if output_styles_dir.is_dir():
             print("  Would install: output styles to ~/.claude/output-styles/ + set outputStyle in settings.json")
+        print("  Would set: env.CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS=1 in settings.json")
         return
 
     _copy_hook_scripts(claude_dir, hooks_scripts_dir)
@@ -41,6 +42,7 @@ def install_hooks(claude_dir: Path, hooks_scripts_dir: Path,
         print("  Removed: .claude/hooks.json (legacy)")
 
     _install_output_styles(claude_dir)
+    _install_env_vars(claude_dir)
 
 
 def _copy_hook_scripts(claude_dir: Path, hooks_scripts_dir: Path) -> None:
@@ -88,3 +90,31 @@ def _install_output_styles(claude_dir: Path) -> None:
             with open(settings_path, "w", encoding="utf-8") as f:
                 json.dump(settings_data, f, indent=4)
             print("  Set: outputStyle = 'Golden Rules' in settings.json")
+
+
+# Toolkit env vars injected into settings.json
+TOOLKIT_ENV_VARS = {
+    "CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS": "1",
+}
+
+
+def _install_env_vars(claude_dir: Path) -> None:
+    """Ensure toolkit env vars are set in settings.json."""
+    settings_path = claude_dir / "settings.json"
+    if settings_path.is_file():
+        with open(settings_path, encoding="utf-8") as f:
+            settings_data = json.load(f)
+    else:
+        settings_data = {}
+
+    env = settings_data.setdefault("env", {})
+    changed = False
+    for key, value in TOOLKIT_ENV_VARS.items():
+        if env.get(key) != value:
+            env[key] = value
+            changed = True
+
+    if changed:
+        with open(settings_path, "w", encoding="utf-8") as f:
+            json.dump(settings_data, f, indent=4)
+        print("  Set: env vars in settings.json (CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS=1)")